1. Field of the Invention
The present invention relates to a recording and/or reproducing apparatus for recording or reproducing on or from a recording medium digital data, particularly those having a variable length, which have been obtained by compression-coding a time-varying image signal. The present invention also relates to a method of recording or reproducing such digital data on or from the recording medium.
2. Description of Related Art
It has been technically very difficult to record or reproduce an image signal as digital data on or from a recording medium because such an image signal contains a great deal of information. By way of example, in applications where an NTSC composite signal is quantized with eight bits at a sampling frequency of 4 f.sub.sc (f.sub.sc being indicative of the color sub-carrier frequency), real-time recording or reproduction of a time-varying image signal as digital data requires a transfer rate of about 120 Mbps. However, the realization of a recording and/or reproducing system achieving the transfer rate of 120 Mbps depends upon the performance between the recording medium and a recording and/or reproducing head, and is, therefore, technically very difficult. Where information is recorded on or reproduced from the recording medium at a high transfer rate, the use of a plurality of recording and/or reproducing heads is effective in reducing the transfer rate imposed upon each of them, or synchronized operation of a plurality of recording and/or reproducing apparatus is also effective.
Although real-time processing can be achieved in the above-described manner, recording or reproduction of all the time-varying image signal information itself is disadvantageous in that the recording or reproducing period of time required is completely determined by the total capacity of the recording medium. To overcome this problem, some apparatus for recording or reproducing a digitalized time-varying image signal are of a design capable of compression-coding the image signal, for subsequent recording thereof on the recording medium, to the extent that deterioration of information of an original signal does not become conspicuous, and of decoding the compression-coded image signal for reproduction, thereby simplifying the real-time recording or reproduction processing and extending the recording or reproducing period of time.
In compression-coding the image signal, several frames of data are processed as one group, and intraframe compression-coding having a relatively low compression rate is applied to at least one frame of data, while interframe compression-coding having a relatively high compression rate is applied to the remaining frames of data. Although the intraframe compression-coding and decoding are carried out within a single frame, if the compression rate is too high, the SN ratio of the image is considerably reduced. On the other hand, the interframe compression-coding and decoding cannot be carried out within a single frame. However, because the interframe compression-coding makes use of correlation with time among frames, it is possible to make the compression rate relatively high. The combination of the intraframe and interframe compression-coding processings not only offers an increase in total compression rate, but prevents the time-varying image having a violent motion from being reduced in SN ratio.
Where recording or reproduction of the compression-coded data on or from the recording medium is desired, the conventional practice is that the intraframe compression-coded data and the interframe compression-coded data are arrayed in time sequence for each frame, and that the recording or reproduction is carried out by the use of a single recording and/or reproducing head.
In the compression-coding method referred to above, however, the quantity of coded data generally considerably varies depending upon differences in the degree of correlation of the input image signal among the frames or the degree of complexity of individual image frames, thus occasionally greatly changing the data transfer rate.
Conventionally, the information transfer rate between the recording medium and the recording and/or reproducing head depends greatly upon the performance of the recording medium or the like. Accordingly, where the compression rate of the coded data is low, i.e., where the compressed data still have a lot of information, a high information transfer rate is required between the recording medium and the recording and/or reproducing head, making it difficult to achieve the real-time data recording or reproduction. Furthermore, it has hitherto been very difficult to record or reproduce data having a variable length while the transfer rate between the recording medium and the recording and/or reproducing head is being changed momentarily.